The last decade has seen an increase in the development and application of new proteomics technologies. These technologies are beginning to make a significant impact on our understanding of basic and clinical questions in cancer biology. However, in the 20 years that we have been able to take tandem mass spectrometry data of peptides and search it against a sequence database, the general strategy has remained largely the same. Basically, protein mixtures are digested to peptide mixtures, the mixture is analyzed by liquid chromatography-tandem mass spectrometry using data dependent acquisition (DDA), and these spectra are interpreted using database searching. Overtime, our field has refined our methods using this same general approach and when combined with improvements in mass spectrometry instrument hardware, the comprehensiveness and throughput of proteomics experiments has improved greatly. To continue to make an impact, we need to further improve the depth and throughput of our experiments. The ultimate and lofty goal being the detection and quantification of all peptides above the limit of detection across many samples. Peptide intensity measurements are currently not comparable between laboratories and platforms without calibration making each experiment an isolated event with little opportunity for labs to build on existing data. While we have made a lot of progress as a field, we need to continue to improve our methods and we need to alter our strategy now so that we can move further forward in the future. We are proposing the development of new data acquisition and analysis strategies that will improve the systematic analysis of large numbers of samples for large-scale molecular phenotyping. Specifically we plan to 1) improve the selectivity of data independent acquisition (DIA), 2) improve the reproducibility and sampling depth of DDA, and 3) change the analysis of the tandem mass spectrometry data from spectrum-centric to peptide-centric. It is necessary to continue to develop both DDA and DIA data acquisition strategies parallel and that with another few cycles of hardware improvements the two methods may ultimately converge. We will then demonstrate the use of these technologies to show that we can rapidly phenotype the cellular state in response to cancer therapeutics.